21 June 2021

TCEC Prepares S21 DivP; CCC Gives Black a Chance

What's new in the world of the two highest level chess engine competitions? To know what's new, we first have to know what's old. Here's a summary of the previous post in the series, TCEC S21 L2 Underway; CCC Unorthodox Openings (June 2021):-

TCEC: Chessdom.com issued two reports on L4. Minic won L3, a half point ahead of Booot. L2 has finished a quarter of the scheduled games. • CCC: After the thematic events, the site continued with 'Belgian Stew', with randomly selected openings; and 'Saragossa', starting 1.c3. The site is currently running a 'Double Bongcloud' event. After that we should probably brace for more unorthodox openings.

That was then. What about now?

TCEC: There have been no further reports from Chessdom.com since the L4 (League 4) results. L2, with 12 engines, was won by Pedone, 1.5 points ahead of Nemorino, the top engine in a field of five separated by a single point. L1, with 8 engines, was won by Igel, 0.5 points ahead of Ethereal which was 1.5 points ahead of the other engines.

The site is currently conducting tests for its Premier Division, 'DivP' in TCEC jargon. The top two DivP engines will meet in the S21 final match, 'Superfinal' or 'Sufi' in TCEC jargon.

CCC: Are we braced for more unorthodox openings? Yes, we are. After the 'Double Bongcloud' event -- does anyone care about the results? (I sure don't) -- the site ran an event called 'The Hillbilly Attack', 504 games starting 1.e4 c6 2.Bc4. For more about the variation, see The Birth and Development of the Hillbilly Attack (chess.com) by GM Simon Williams.

Why don't I care about the 'Double Bongcloud' (1.e4 e5 2.Ke2 Ke7)? Because it's an artificial construction that will only occur by force or by collusion between the two opponents. Any player of the Black pieces possessing free will would continue with some logical developing move after 2.Ke2. It could be the junkiest of the junk openings. In contrast, 'The Hillbilly Attack' has some theoretical interest. Being rooted in the Caro Kann (1.e4 c6), it's a legitimate attempt to explore one of the major defenses to 1.e4.

The site is now running an event called 'Mini-Match: Leela vs Dragon, Muzio Gambit', 200 games starting 1.e4 e5 2.f4 exf4 3.Nf3 g5 4.Bc4 g4 5.O-O gxf3. That last move, 5...gxf3, doesn't appear to be part of the forced moves, but I'd like to see anyone not accept the Knight sacrifice. While I was writing this paragraph, a revealing exchange scrolled through the chat:-

PG: I just realised that White never won a game in this tournament. • PL: People at Chess.com got tired of White always winning in engine games so they picked openings that give Black an advantage.

Anyone interested in a course on chess logic? Maybe the CCC should investigate White giving various odds to Black. It might even advance chess theory in ways we can't imagine.

[For further information from the various stakeholders in the engine-to-engine events, see the tab 'TCEC/CCC Links' at the top of this page. • NB: Leela = LC0 = LCzero]

No comments: